Algorithm-based design optimization for building material reuse: Integrated path generation and reclaimed stock assignment

Abstract

Reuse is not commonly adopted in practice, despite its acknowledged benefits, partly due to the complexity of the design process, considering geometric constraints and fluctuating stock availability. A multi-objective optimization framework for algorithm-based stock assignment and path generation is developed, specifically for one-dimensional material systems (e.g., piping, timber, steel), to maximize reuse allowing for serial connections of stocks. Under the overall genetic algorithm-inspired optimization structure, improved A* and heuristic algorithms are used for pathfinding and stock assignment, respectively. A comparative analysis of the fitness values for both lab-based and real-world case studies demonstrates the robustness of the approach with different optimization parameters, stock availability, and scale complexity. This method can help the next users of construction and demolition waste and spare maintenance parts reuse more materials, contributing to a circular economy in the construction industry. Future research can further expand and apply the proposed approach to more complex real-world scenarios.